A case of the Hum being a tale of two sub-stations,

by Dr Chris Barnes Bangor Scientific and Educational Consultants, April 2013.

Abstract

The Hum is defined and what is known about it briefly reviewed. Copies of acoustic measurement spectra printouts have been provided to the author said copies were obtained by a local authority in relation to a noise complaint associated with the Hum in Charlbury. A frequency analysis of these measurements has been made. It is shown that the greater part of the acoustic spectrum present in the complainant’s premises can be accounted for by the spectrum recorded at two nearby sub-stations and under an electricity pylon outside the complainant’s premises. An unusually high acoustic signal at 100 Hz is found within the premises with the electricity supply to the premises switched off. Possible reasons for this are discussed. The great majority of frequencies present at the premises are also features of the Bangor Hum some 400 km distant. A specific frequency component of the spectrum at 56 Hz is identified as it has also been found in relation to a Hum in New Zealand. The consensus is that Hums may arise as a consequence of interaction between Electricity and Gas infrastructure particularly when electricity carries excess harmonic, sub-harmonic, and inter-harmonic with associated phase imbalance and stray ground currents due mainly to wind energy. Nevertheless the injection of external frequencies such as those transmitted by ionosphere heaters into the Alfven resonator cannot entirely be ruled out as these may be received by power lines or gas grids acting as long antennas.

Introduction

The Hum is a term given to a modern and relatively unexplained LFN phenomenon (1). The first academic publication on the Hum, Deming, blamed TCAMO military communications aircraft as having the most likely correlation in time and space with the Hum (2). Those afflicted describe hearing a distinct, yet annoyingly penetrating, sound akin to a slowly, irregularly idling engine. Others describe a noise like a giant bee trapped in a bottle perceiving the noise behind the ear or head rather than inside the ear. Musically talented individuals tend to tone match the Hum at between 30-80 Hz and state that it has pseudo-random modulations in frequency and amplitude of between .5 and 5 Hz. The phenomenon is also described as difficult or almost impossible to screen with standard earplugs and often worse in the dead of night. The Hum began in the UK in the 1970’s when gas mains and motorways, new infrastructure of the era, were blamed. The only other new infrastructure installed at that time was electricity infrastructure in the form of pumped storage power stations. Nor did anyone ever consider potential interactions between signals form different infrastructure as arising the Hum. Amateur Hum investigator John Dawes has long since blamed electricity infrastructure for the Hum yet until recently without academic back up. The Hum was not heard in the USA until the 1990’s wherein, perhaps because in Taos NM it could not readily be audio recorded by Mullins(3), although he did suggest that some hearers ears might be more acute than the existing equipment thus it was regarded by some as an almost supernatural phenomenon. Occasionally some cases of the Hum have been traced directly to natural sources such as volcanic action (1) or to industrial sources such as compressors and fans (4). As a scientist and some years ago and at the onset of being asked to investigate UK Hums, the present author viewed the Hum as being either a preternatural or an anthropogenic phenomenon yet to be explained. If one looks at all the available anecdotal knowledge on the Hum one begins to realise that it far more than just a noise. Medical acoustician David Baguley has blamed the Hum on over-sensitive hearing(5) but the present author believes this to be an oversimplification and such a hypothesis does not explain an onset time for the Hum nor does it explain why more and more people’s hearing is suddenly becoming ‘over-sensitive’.

The present author has spent some eight years or so investigating UK Hums and has also been in contact with US, Canadian and New Zealand Hum investigators.

Being a subjective phenomenon doing science on the Hum has been difficult and time consuming and one has had to rely on many anecdotal reports and one has had to try and evolve one’s theories to satisfy all that is known or stated. What has helped, however, is that the author’s wife has perceived the Hum since early 2004 and the author started to perceive it about a year later. It is reckoned that estimates of between 2-11% of the world’s population perceive the Hum and that Hum perception maximises in middle age (2).

The present consensus is that the main route for perception is the presence of infrasound and low frequency acoustic sound in a highly skewed acoustic spectrum with relatively silence above about 200Hz. In some individuals there appears to be one or more synergetic mechanisms of perception at work in addition to this main route which are either bio-electromagnetic or bio-gravitational or both (6,7).

At least for the Hum perceived in Bangor, Gwynedd, the author has shown a link to the electricity supply industry (8), but multiple sources may be required for Hum perception (9,10).

For world Hums in general there has been shown a recent strong geo-spatial correlation with renewable energy systems especially wind power (11). The link is however not thought to be simply or entirely due to direct reception of airborne turbine noise in premises.

Indeed Hum levels both in Bangor and as reported elsewhere appear to wax and wane and are apparently dependent on many earth bound and even space physics linked variables (12,13). Thus to fully explain the Hum some understanding of earth science and space physics has to be gleaned.

It is very rare for a Hum investigator to be able to hear Hums in premises. What is reported on here then is perhaps almost unique. The present study reports on acoustic spectra recorded in premises in Charlbury, Oxfordshire wherein a Hum was actually perceived not only by the householder but also by the environmental health officer.

The householder was adamant that the Hum was somehow associated with electricity even though it could be heard with the electricity supply to the building cut off. The assumption was made on the basis that the complainant perceived the same or similar noise outside electricity sub-stations or underneath power-lines. Thus the environmental officer was persuaded to make measurements on the boundary of two nearby electricity sub-stations. The complainant was also adamant that some of the noise may originate from an electricity pole outside the premises, thus measurements were made under this as well.

Experimental

Experiments were conducted on the evening of 18th December 2012 at circ 2200 GMT at the address in Charlbury and the neighbouring sub-stations. The apparatus was a battery powered portable FFT based Brüel & Kjær narrowband Sound & Vibration Measurement Meter. The system works by collecting average FFT data over about 30 seconds thus unlike the colour waterfall output system used by the present author elsewhere it is impossible to tell if a particularly spectral component is static or pulsating in amplitude. The spectra recoded were later downloaded for print out.

Results and Discussion

Figure 1

Figure 1 above shows the acoustic spectrum recorded in the bathroom of the premises affected by the Hum. The most apparent feature is the highly skewed spectrum with no frequencies present above 200 Hz. Below this there is comb -like spectrum with major peaks 127 Hz, 100 Hz, 23 Hz, 21 Hz and approximately 1.5 Hz. There are other lesser peaks at 6,12,28.4,35.8,38.3,41.9,44,47,50,55.8,67,70,77.3,91.2 ,94,108,111 and 121 Hz. It is interesting to note that the original Bristol Hum was tone matched to 36 Hz and that is extremely close to one of the spectral components found here.

Figure 2 shows the spectrum 1 metre distant from the entrance of the nearby Sheep Street sub-station there are a number of major, narrow tonal components extending from about 1.4 Hz to 300 Hz with significant maxima at 1.4, 23,35, 94 and 200 Hz and other notable spikes at 12,13,31,36,47,56,60,84,117,128 and 300 Hz. Components of 31 and 36 Hz have been mentioned elsewhere by Sergeant and by Leventhall in relation to the Hum.

Figure 3 above shows the spectrum at the other nearby sub-station, namely Woodfield drive. There is tonal noise and across the entire spectrum with harmonic peaks to 450 Hz and broad band noise below 150 Hz. The most significant tonal peaks are as is expected of a typical substation i.e. 100 Hz and also at 200,300,400 and 450 Hz. Within the broader band section there are more distinct signals at 1.6,23.7,28.3,50,70.4,94,121 and 150 Hz.

Figure 4 above shows the spectrum recorded beneath the pole adjacent to the premises.

Broadband noise between 0-125 Hz was apparent. Most significant peaks were at 2.3,4.6,6.7, 10,16.8,28,41.8,56,60,75,100,110,127 Hz.

What is evident is that 80 % of the major acoustic and infrasonic frequencies present at the premises can be provided by the two substations and also some 40% of those with lower amplitudes. Those provided by the Sheep Street substation have been highlighted in yellow and those provided by the Woodfield drive substation have been highlighted in green. More speculative is the broader infrasound feature at 1.5 Hz highlighted in purple. It is conceivable this could arise as overlap between the broad feature at 1.6 Hz at Woodfield drive and the narrower feature at 1.4 Hz provided by Sheep Street. Sub-stations have been reported to produce pulsation when transformers operate at or close to magnetic saturation noise this is a particularly aggravating feature of Hums. Other sources of such pulsation can be power flow oscillations and wind turbine flicker. Since these features are quite broad, it is difficult to compute exact frequencies from the FFT spectrum. There is even some overlap into the 2 Hz band. A ubiquitous 2 Hz signal has been reported in parts of Northern Europe and Asia as being related to Hydropower.

No feature seems present at the premises higher in frequency than circa 128 Hz. It is of course well known that low acoustic and seismic frequencies propagate though air and ground respectively with least attenuation (14.15). Overall electricity external infrastructure including the two substations and the pole would seem to be able to account for about 70% of the entire acoustic spectrum present in the premises including several key frequencies which have been shown to be associated with other Hums in geographic locations significantly elsewhere, in Wales for example. The background broadband spectrum encompassing the infrasonic range and LFN range up to about 150 Hz under the pole looked most like that in the premises. This does not necessarily prove that electricity infrastructure is responsible for this background spectrum it may just be that the pole was acting as an amplifier and re-radiator of ground borne vibration (16). The author has stated elsewhere that people who perceive the Hum (hummers) appear to be very sensitive to natural vibration as well as anthropogenic noise and indeed this may be a survival instinct as a great deal of people interviewed who hear the Hum have previously experienced and lived through earth tremors and earthquakes(17).

The most striking single result was that the amplitude of the 100 Hz tonal component at the premises of 40dB was the same as that only 3m away from the entrance at the Woodfield Drive sub-station. This substation is itself some considerable distance away from the afflicted premises. The other frequencies at the premises may arise as a result of either signals due to other utilities or as a result of non-linear mixing. The author has commented on these effects elsewhere (18). The complainant describes the Hum at the premises in the classical manner that it has quasi-periodic fluctuations. Unlike the waterfall display analysis system used by the author elsewhere, the equipment used by the investigating local authority did not have this facility and so it was not possible to tell if the 100 Hz component or any of the other components were fluctuating significantly in a short space of time.

In this particular case the author most certainly believes further investigation is warranted. The mechanism by which the sound, particularly the 100 Hz tonal component, arrives at the premises totally without apparent path loss is indeed intriguing. A noise level of 40dB at night is an unreasonable in a normally very quiet rural area and sufficient to delay sleep onset or cause sleep disturbance (19). It may be that there is a room or room resonances close to the offending frequency, such concepts have been described by the author elsewhere (20). It could be that some magnetically permeable material within or attached to the building fabric lies within the induction field of the outside pylon and is vibrating accordingly (21). It could be that the premises are built on piezoelectric or magnetic rock through which a substantial ground current is flowing thus inducing vibration of the premises as a whole(22). Madshus et al 2005(23) could provide an alternative intriguing answer to this if there is present both an airborne acoustic wave and a related ground borne visco-elastic Rayleigh Wave. They found that the pressure wave interaction with the viscoelastic Rayleigh wave in the ground may have a. significant effect on the ground impedance and the sound and vibration propagation. This introduces an important mechanism not covered in commonly used ground impedance models. They developed numerical simulation models and verified against the test data. The ground impedance does not only effect the sound pressure propagation. If either acoustically induced ground vibration, or ground to building transmitted vibration, is to be considered, the acousto-seismic impedance has a dramatic effect on the level of ground vibration induced by a given sound pressure. For a site where Rayleigh wave interaction appears at the dominant frequencies of the sound pressure, the ground vibration may be greater than a factor 100 (40 dB) than at a site with ground conditions not making the interaction happen. Such might be exactly the example with, for instance an electricity substation with ground mounted transformers. It is interesting to note that Moir (24) has observed that for the Hum in New Zealand this appears to maximise in houses in a depressions or hollows somewhat lower than adjacent roads. Intuitively, one would feel this might be the ideal condition for a surface Rayleigh Wave and an airborne wave with some facet of coherence to reinforce. In Bangor the Hum situation could be similar but will be complicated by the fact that a ground borne signal at 8.3 Hz due to the Dinorwig pumped storage system is also present.

What is clear in Charlbury is that the complainant noticed a distinct commencement of the phenomenon in July 2010. At about this time changes commenced in local railway infrastructure, it is unclear if this would have affected either electrical ground currents in any significant way or the propagation of ground borne seismic signals. What is also clear is that 2009 -2010 saw the greatest year on year increase in wind power generation in the UK (25). and this would be expected to impact on power quality and power flows in the Charlbury region as across the UK generally.

Similarities between Charlbury Hum and Bangor area Hums

To re-iterate, the frequencies noted in the afflicted premises in Charlbury were 127 Hz, 100 Hz, 23 Hz, 21 Hz and approximately 1.5 Hz. There are other lesser peaks at 6,12,28.4,35.8,38.3,41.9,44,47,50,55.8,67,70,77.3,91.2 ,94,108,111 and 121 Hz, and broadband background infrasound with relative silence above 150 Hz.

The frequencies noted in the author’s house from several studies are: 122,100,88,65,64,60,56,50,46,33,31,28, 8.3 and 4 Hz.

What is apparent is that there are striking similarities between the two sets of spectra. This is quite remarkable given two very different premises in different parts of the U K. Closest similarities have been highlighted in yellow and those up to 2 Hz different in grey. Clearly the electricity infrastructure is capable of providing acoustic frequencies of 50 and 100 Hz into people’s premises as discussed here and elsewhere and from the signals at recorded adjacent to the two sub-stations would appear capable potentially of providing a whole load of other frequencies in addition, which in Charlbury were: 1.4, 1.6, 4,6,12, 13, 23, 28,31,33,35, 47, 56, 60, 64, 84,100, 117,121,128,200,300,400 and 450 Hz. The only low (infrasound) frequency not experienced in Charlbury but present in Bangor is that at 8.3 Hz thought to be associated with the Dinrowig pumped storage facility. Clearly the signal propagation path is limited in that direction. Whereas seismic signals of lower than this frequency would appear to propagate significantly further.

In other words virtually all the frequencies experienced some 400 km distant in the author’s home.   It is instructive to compare the result for the Anglesey double circuit where there is a wide band of infrasound at 5-12 Hz there are also unexplained signals at 23,36,56 and 112 Hz. Peak sound levels 78dB(A) (16). http://www.drchrisbarnes.co.uk/UTILITY.htm

The question now arises do any of the frequencies in the complaint’s premises not arise as a result of electricity infrastructure? The 111 Hz peak in the bathroom at Charlbury does seem disproportionately high compared with that part of the spectrum at either of the substations but it is interesting to note that the power grid in Anglesey does appear to provide almost that frequency i.e. 112 Hz. It would seem that frequencies of 21Hz, 38.3 Hz, 44 Hz, 67 Hz, 77.3 Hz, 91.2 Hz and 108 Hz found in the house in Charlbury are not supplied directly by the electricity infrastructure. There are two choices here. Either they arise because of non-linear mixing processes or they are supplied by other infrastructure.

Four out of seven of these frequencies to within 1 Hz or so, also highlighted in green have also been observed in the Bangor and Anglesey area in association with High Pressure Gas mains and Gas Pressure Reducing Stations (16). http://www.drchrisbarnes.co.uk/UTILITY.htm.

Most perplexing of all is the frequency of 56 Hz which has been recorded in association with both gas and electricity infrastructure and which has also been audio recorded in New Zealand as in association with the Hum(24).

Many of these frequencies also correspond with or are very close to natural Earth Schumann Resonances (26) to which it is argued human brain rhythms have also evolved (27). This might account for why Hums are so difficult to screen out because the brain is trying to process information from them. There is of course also evidence that the HAARP system in Alaska is capable of transmitting on these frequencies (28). Small wonder then by wind borne mains frequency inter-harmonic, gas compressor, or powerful transmitter in Alaska or a combination of the three more and more people throughout the world are being plagued by the Hum.

Conclusions

This study is extremely important because the results it reveals appear to reinforce much of the author's previous work (29) yet the data has been acquired by a totally independent body and by equipment other than that normally used by the author. The complainant states that the Hum in question is audible at other locations in Charlbury. This is just as the hum in Bangor is audible across parts of and premises in the surrounding area. Back in the 1970’s Vasudevan and Gordon (30) stated that they thought the Hum was virtually everywhere. A good friend of the complainant also hears the Hum and has stated that the only time she gets relieve from the Hum is when she visits a part of Texas in the USA which has neither gas nor wind power. So perhaps the Hum is everywhere in Britain. The Hum has grown at an alarming rate since the 1970’s with most new cases in Britain coming in the last eleven years or so. Gas infrastructure has stayed more or less the same with the exception of two major new pipelines in Wales. Electricity infrastructure has changed beyond recognition with the inception of smart grids and vastly expanded wind power. There is thus a potential pointer that the Hum, a facet of modern living, may be both gas and electric.

It has been shown that most, possibly all of the frequencies found in premises wherein occupants suffer from the Hum can be potentially supplied by electricity infrastructure. An unusually high level of 100Hz audio exists in the complainant’s premises which cannot be explained by routine acoustic propagation. This may be explainable because of room resonance, the presence of magneto-strictive foundations to the premises or such material within or attached thereto and/or the Madshus et al 2005 may be very relevant to this and other Hums worldwide. Clearly further investigation is required. This strong 100 Hz signal exists at the premises together with almost a comb spectrum of other narrow band signals of varying amplitude some possibly at amplitude levels traditionally regarded as too low to cause annoyance but this would change if they were modulating features and these are super-imposed on a broad band infrasound background in an acoustically skewed spectrum pretty much quiet above 200 Hz and shown on several occasions elsewhere to be requisite for or feature of the Hum (30,31). Due to the equipment employed it has not been possible to tell, if for instance the 100 Hz signal was amplitude or frequency modulated by any of the lower frequency features observed. The complainant describes the noise as modulated in the typical sense of the Hum as described by the majority of the afflicted. Given that there is infrasound present both broad band and tonal in addition to the 100 Hz, this is hardly surprising. Modulation effects can occur due to non –linearity in the ear, see Moller and Pederson 2004 (32) and this helps explain why the few hums which have been audio recorded rarely, if ever, mirror that which a complainant perceives.

In Britain, high pressure gas mains often share the same land corridors as those used by overhead electricity supply at both National Grid and DNO level (33). Due to induction common frequencies will appear. As more and more non –linear loads and cyclo-converters are used in the electricity industry, steel and cement industry and railways it is hardly surprising that more and more inter-harmonic is turning up and in turn affecting even the acoustic vibrations of sub-station transformers. Interestingly cyclo-converters were first used in Britain in the 1970’s (34) the time the Hum first turned up. Recently sub –harmonic groups ( likely to be more relevant for the Hum) have been shown to be stronger than expected on low voltage circuits, see Barros 2007 (35). Precise methods of calculation accounting for fundamental frequency drift are given by Vahedi et al, 2013 (36). It is at present impossible to say if, for example, the 56 Hz frequency is an inter-harmonic or is associated with gas compressors. Also, both systems have extensive pipe and cable runs, many thousands of kilometers around Britain and will therefore potentially make receiving antennas for external signals such as natural Q bursts within the Schumann resonance bands or for ELF communication signals which may have been injected into the Schumann cavity. These electromagnetic signals can get also transformed into acoustic signals in number of ways; this further complicates the issue of the Hum.

References

1. http://en.wikipedia.org/wiki/The_Hum

2. http://www.scientificexploration.org/journal/jse_18_4_deming.pdf

3. http://www.johndawes.pwp.blueyonder.co.uk/taos.htm

4. http://www.johndawes.pwp.blueyonder.co.uk/kokomo1.pdf

5. http://news.bbc.co.uk/1/hi/8056284.stm

6. http://www.drchrisbarnes.co.uk/HUMCAVE.htm

7. http://www.drchrisbarnes.co.uk/HUM3FIELD.htm

8. http://www.drchrisbarnes.co.uk/BANGORhum.htm

9. http://www.drchrisbarnes.co.uk/UTILITY.htm

10. http://www.drchrisbarnes.co.uk/HUMHOUSE.htm

11. http://drchrisbarnes.co.uk/HUMGRIDNEW.htm

12. http://www.drchrisbarnes.co.uk/geomag.htm

13. http://www.drchrisbarnes.co.uk/HUMIMF.html

14. http://www.sfu.ca/sonic-studio/handbook/Sound_Propagation.html

15. http://en.wikipedia.org/wiki/Seismic_communication

16. http://www.sciencedirect.com/science/article/pii/S0022460X03009246

17. Personal communications with affected subjects

18. http://www.drchrisbarnes.co.uk/HUMROOM.htm

19. www.euro.who.int/document/e92845.pdf

20. www.drchrisbarnes.co.uk/HUMROOM.htm

21. Development near overhead lines - National Gridwww.nationalgrid.com/NR/.../Developmentnearoverheadlines.pdf

22. http://www.drchrisbarnes.co.uk/TheHumQuestionsandAnswers.htm

23. http://www.sciencedirect.com/science/article/pii/S0003682X04001598 Air-ground interaction in long range propagation of low frequency sound and vibration - field tests and model verification

Madshus, C.; Lovholt, F.; Kaynia, A.; Hole, L. R.; Attenborough, K. and Taherzadeh, S. (2005). Air-ground interaction in long range propagation of low frequency sound and vibration - field tests and model verification.Applied Acoustics, 66(5), pp. 553–578.

24. http://www.speechresearch.co.nz/humspec.jpg

25. http://en.wikipedia.org/wiki/Wind_power_in_the_United_Kingdom

26. http://www.glcoherence.org/monitoring-system/earth-rhythms.html

27. http://researcharchive.lincoln.ac.nz/dspace/bitstream/10182/3935/1/90_n1_EMR_Schumann_Resonance_paper_1.pdf

28. http://onlinelibrary.wiley.com/doi/10.1029/1998RS900014/abstract;jsessionid=ECCACBF8C87E1FD332668F3DC44C8E07.d04t04?deniedAccessCustomisedMessage=&userIsAuthenticated=false

29. http://www.drchrisbarnes.co.uk/HUMPUBS.htm

30. http://www.drchrisbarnes.co.uk/Silence.htm

31. http://books.google.co.uk/books?id=b3Xqfc2KRo8C&pg=PA27&dq=Vasudevan+and+Gordon++Southampton&hl=en&sa=X&ei=SI59UYrLHMHBPJvVgLgB&ved=0CDYQ6AEwAA#v=onepage&q=Vasudevan%20and%20Gordon%20%20Southampton&f=false

32. H. Moller and C.S. Perderson, ‘Hearing at low and infrasonic frequencies’, Journal of Noise and Health 2004, 6 (23) pp37-57.

33. www.elec.uow.edu.au/iepqrc/content/papers/AUPEC/AUPEC04_6.pdf

34. Interharmonic Task Force Working Document IEEE 2001

35. J.Barros, Measurement of Sub-harmonics in power voltages, Conference Proceedings, Power Tech 2007, IEEE, Lausanne.

36. Vahedi et al, Journal of Power Electronics Vol 13 (1), Jan 2013.